Me: Can we have a framework performance test?
Mom: we have framework performance test at home
Framework performance test at home:
Because I wanted to understand the difference between the frameworks and how they perform in a real world scenario.
Plus I had this very specific use case to fetch a single int from database and send it to the client. I started from the tech empower benchmarks but I didn't understand a lot of things about their testing method and setup so I decided to do my own.
- t2.micro EC2 instance as client in eu-central-1a
- t2.micro EC2 instance as server in eu-central-1a
- db.t3.micro RDS instance in eu-central-1a
I have also ran the tests on a t3a.xlarge EC2 instance in eu-central-1a and the results were pretty much the same.
I had ran the tests with and without nginx. As it turns out nginx doesn't get the processing power from the astral plane and its resource usage affected the results.
2 endpoints:
/- hello world as plain text/count- fetches a single int from database and sends it to the client as plain text
- rewrk is used as a testing tool:
rewrk -c 50 -t 2 -d 30s -r 5 -h http://localhost...is the command to run the tests - on edge the
./build.shand the./bench.shscripts were used
Every test was ran 5 times and the average of the 5 runs was used to compensate for the changes in the environmental conditions.
The output from the last run can be found in the raw_output.txt
https://docs.google.com/spreadsheets/d/1qxV0QYc6v51SDrhph7MY6PEOdoGL_IYRNgvRJmV1GKk/edit#gid=0
